Table 2. Accuracy metrics of automated assessment models classifying reptile species into IUCN extinction risk categories, under 2 different approaches: (1) complete model, accounting for spatial and phylogenetic autocorrelation and assessor/reviewer effects; (2) accounting for spatial and phylogenetic autocorrelation (this was the model used for predictions).

	Binary	CR	EN	VU	NT	LC
Complete
Sensitivity	0.955	0.773	0.699	0.532	0.278	0.964
Specificity	0.711	0.997	0.976	0.977	0.983	0.691
AUC	0.833	0.885	0.837	0.755	0.631	0.828
Precision	0.925	0.927	0.731	0.649	0.556	0.890
Recall	0.955	0.773	0.699	0.532	0.278	0.964
F1	0.940	0.843	0.715	0.585	0.370	0.925
Spatial-phylogenetic (used for predictions)
Sensitivity	0.952	0.621	0.726	0.278	0.532	0.950
Specificity	0.703	0.995	0.969	0.976	0.979	0.683
AUC	0.828	0.808	0.847	0.627	0.756	0.816
Precision	0.923	0.872	0.689	0.463	0.667	0.886
Recall	0.952	0.621	0.726	0.278	0.532	0.950
F1	0.938	0.726	0.707	0.347	0.592	0.917

“Binary” represents the separation of threatened (CR, EN, and VU) from nonthreatened categories (NT and LC). Remaining columns represent the predictive accuracy for assigning species to the 5 extinction risk categories: CR, Critically Endangered; EN, Endangered; LC, Least Concern; NT, Near Threatened; VU, Vulnerable. See S1 Table for remaining models.